The two-segment caliber method of diagnosing iliac vein stenosis on routine computed tomography with contrast enhancement.

Background: Iliac vein stenosis is a frequent pathologic process in advanced chronic venous disease. Intravascular ultrasound (IVUS) has emerged as the "gold standard" to diagnose iliac vein stenosis and to guide stent treatment. A pre-IVUS test is often obtained. Routine venography is deficient in several respects to fill this role; absence of an internal scale is a critical deficiency. Computed tomography venography (CTV) may be superior; its measurement capabilities can be used to precisely identify stenotic iliac vein caliber. Furthermore, the calibers of common iliac vein (CIV) and external iliac vein (EIV) can be individually assessed, yielding two data points instead of single-point assessment used in venography and current CTV protocols. We compared the diagnostic accuracy of the two-segment caliber method of CTV (arm vein injection of contrast material) with IVUS.

Methods: In patients who underwent computed tomography assessment of iliac vein segments before IVUS examination during a 5-year period, 91 limbs were analyzed. This is a single-center, retrospective analysis of prospectively collected data. CTV images of CIV and EIV segments were compared individually and in combination with IVUS planimetry images. A caliber diameter of <16 mm for CIV and <14 mm for EIV was considered stenotic with either imaging technique. Bland-Altman plots and receiver operating characteristic curves were used.

Results: On IVUS evaluation, 84% of CIVs and 78% of EIVs were stenotic and 16% and 22% were of normal caliber. These provided IVUS positive and negative controls for CTV comparison. On CTV, at least one of the two segments (CIV or EIV) was stenotic in 90% of the limbs, about 10% to 15% higher than single-segment involvement. Mean CTV caliber difference from IVUS was +2.5% for CIV and +7.3% for EIV. On Bland-Altman plot, single-segment diagnostic sensitivity of CTV was 83% and 73% for CIV and EIV, respectively, compared with IVUS. The sensitivity increased to 97% with a positive predictive value and accuracy of 93% and 91%, respectively, when a stenotic caliber in at least one of the two segments was considered diagnostic of iliac vein stenosis. Receiver operating characteristic analysis confirmed increased accuracy of the two-segment method over single-segment assessment with an area under the curve of 0.89 (P < .001).

Conclusions: Caliber diameter of <16 mm for CIV or <14 mm for EIV on routine CTV imaging appears to correlate with IVUS caliber stenosis with good diagnostic metrics of low false positives and false negatives.